CheckRLM: Effective Knowledge-Thought Coherence Checking in Retrieval-Augmented Reasoning

2026-07-02

AI Summary

Current reasoning language models are prone to factual errors in knowledge-intensive tasks, which can disrupt the entire reasoning chain. This work proposes the first mechanism capable of detecting and correcting knowledge inconsistencies in real time during inference. By integrating a retrieval-augmented generation framework, the approach dynamically extracts factual claims, precisely identifies inaccuracies, and injects external knowledge with minimal intervention to restore consistency between knowledge and reasoning. The method substantially mitigates error propagation in long-chain reasoning, outperforming existing baselines across multiple knowledge-intensive benchmarks while achieving higher accuracy and reliability at reduced computational cost.

Abstract

Reasoning Language Models (RLMs) have significantly improved performance on complex tasks by extending the reasoning chain. However, these chains are prone to containing factual errors, particularly in knowledge-intensive tasks. To address this issue, we propose CheckRLM, a framework that improves the reliability of the reasoning process through Retrieval-Augmented Generation (RAG) by timely checking and correcting factual errors. Specifically, CheckRLM extracts factual claims from the reasoning chain to identify and localize subtle knowledge inconsistencies during inference. Upon detection of errors, a refinement mechanism performs minimal-cost yet precise corrections by leveraging external knowledge, ensuring coherence between the reasoning chain and correct knowledge. Extensive experiments demonstrate that CheckRLM substantially outperforms existing baselines, exhibiting a strong capability to mitigate error accumulation in long-horizon reasoning with lower costs. The code and data are available at https://github.com/AI9Stars/CheckRLM.

Problem

• Reasoning Language Models
• factual errors
• knowledge-intensive tasks
• reasoning chain
• error accumulation

Innovation

• Retrieval-Augmented Reasoning
• Fact-Checking
• Reasoning Language Models
• Knowledge Coherence
• Error Correction